CFD simulation of gas-liquid stirred vessel: VC, S33, and L33 flow regimes

• Published in: AIChE journal Vol. 52; no. 5; pp. 1654 - 1672
• Main Authors: Khopkar, Avinash R., Ranade, Vivek V.
• Format: Journal Article
• Language: English
• Published: Hoboken Wiley Subscription Services, Inc., A Wiley Company 01.05.2006; Wiley Subscription Services; American Institute of Chemical Engineers
• Subjects: Applied sciences; Bubbles; Chemical engineering; computational fluid dynamics (CFD); Exact sciences and technology; Experimental data; flow regimes; Fluid dynamics; gas holdup distribution; Gases; Hydrodynamics of contact apparatus; Liquids; Mixing; Rushton turbine; stirred vessel; Turbines; Turbulence models; Turbulent flow; Gas holdup; computational fluid dynamics (CFD); Euler equation; Stirred vessel; Correlation; Turbulence; Two phase flow; Computational fluid dynamics; gas holdup distribution; Interphase; Hydrodynamics; Turbine impeller; Forecast model; Modeling; Rusliton turbine; Bubble; Flow regime; Drag; Correlation analysis; Gas liquid flow; flow regimes; Vortex
• ISSN: 0001-1541; 1547-5905
• DOI: 10.1002/aic.10762

A comprehensive computational model based on the Eulerian–Eulerian approach was developed to simulate gas–liquid flows in a stirred vessel. A separate submodel was developed to quantitatively understand the influence of turbulence and presence of neighboring bubbles on drag acting on bubbles. This submodel was used to identify an appropriate correlation for estimating the interphase drag force. The standard k–ϵ turbulence model was used to simulate turbulent gas–liquid flows in a stirred vessel. A computational snapshot approach was used to simulate motion of the standard Rushton turbine in a fully baffled vessel. The computational model was mapped onto FLUENT4.5, a commercial CFD solver. The model predictions were compared with the previously published experimental data of Bombac and co‐workers. The model was used to simulate three distinct flow regimes in gas–liquid stirred vessels: vortex clinging (VC), alternating small cavities (S33), and alternating large cavities (L33). The predicted results show reasonably good agreement with the experimental data for all three regimes. The computational model and results discussed in this work would be useful for understanding and simulating gas holdup distribution and flow regimes in stirred vessels. © 2006 American Institute of Chemical Engineers AIChE J, 2006